Working with the cube today. When I power up, the speaker "clicks" and the drive/fan momentary comes on, then everything shuts off. Is this likely a power supply or motherboard issue? I examined both and nothing is visibly damaged. Is there a way to test the PS on it's own?
A NeXTstation has a battery on the mainboard. When it is empty, the computer will not power up. Maybe this can also happen with a cube, but I'm not sure.
I don't believe that a weak battery could cause the PSU to shut down, but it can definitely prevent it from powering on.
It sounds like an overload protection circuit cutting in and shutting the power supply down. Unfortunately I don't know if there are any schematics available for the cube PSU. I would venture a guess that it is most likely the PSU, although an overload somewhere on the main board could be causing an excessive current draw too.
It has been 16 years since I sold my cube and I cannot remember if the PSU label has any ratings printed on it. With a pinout of the PSU connector and the current ratings you could create a simulated load (resistive) and try to trigger it on. But you will need some equipment to do additional testing.
On some of the power supplies (older ones), if you don't draw enough, the power supply will shut down. I've seen that happen when moving from an older drive to a newer one, or using a cube without the monitor.
Does anyone happen to have a pinout for the power supply connector?
For starters, unplug the power AND data from the MO drive. In the past it has caused cubes to not power on.
Same with the SCSI drive for testing purposes.
All, I have disconnected everything but the MB and PS with the same result. I also replaced the battery and still had no luck.
In machines this old, capacitor failures are rampant and can cause a host of power issues. I'd suggest checking the power supply and the main logic board for any signs of capacitor failure, especially the surface mount aluminum cans on the logic board. These fail and leak corrosive electrolyte.
Quote from: "mcdermd"In machines this old, capacitor failures are rampant and can cause a host of power issues. I'd suggest checking the power supply and the main logic board for any signs of capacitor failure, especially the surface mount aluminum cans on the logic board. These fail and leak corrosive electrolyte.
That is a very good point and a real possibility, as I also highlighted in this older thread:
http://www.nextcomputers.org/forums/viewtopic.php?t=2803.Based on barcher174's message I made an assumption, perhaps incorrectly, that his visual inspection did not reveal any anomalies. For reference, here are some photos that show what my leaking slab capacitors looked like:
http://www.flickr.com/photos/t-rexky/6795730212/http://www.flickr.com/photos/t-rexky/6795732116/
Here are some bad ones on a Macintosh LC 475 I repaired recently:
http://i49.tinypic.com/2j315aa.jpg
Thanks guys. Yes, visual inspection did not reveal any bulging or leaking caps. My meter is not advanced enough to check the values in circuit. I have resigned myself to purchasing a known good power supply and then I can continue to debug from there. I very much appreciate your input.
Ok, so new power supply and logic board arrived today. I put them in, and I have the same symptoms. I have tried the soundbox on my nextstation and it works with no issues. I also tried starting the cube by shorting pin 6 to ground and no luck. I am now suspecting the backplane. Looking at it, I see no obvious signs of damage. The only component on it really is the 220ohm resistor network, which tests ok with a meter. Has anyone seen a similar failure?
Quote from: "barcher174"Has anyone seen a similar failure?
Possibly. The piece that is managing the power-on is the time-of-year, real-time-clock (RTC) that also maintains nvram. One of my turbocube boards was failing to boot and evidence pointed to this. The poweron signal (which is pulled to ground by hitting the power key) wasn't reading the full expected voltage (either 3.6 from the battery or 5 from a trickle supply - I forget which). I had a spare cube board which fixed the problem but I've wanted to start looking for replacements for the RTC. This looks like it may be a challenge - think it was a Motorola part - long obsolete.
aha! you mentioned a soundbox, so you don't have a megapixel display connected to the cube? cube will not start if it has the older power-supply built in. somebody correct me if I'm wrong: the old power-supply has model-nr 152.00. you need to get the later model or connect the NeXT mono CRT :!:
EDIT: I actually took apart one of my cubes and what you need is the power-supply with "NeXT Part No: 983" on the sticker.
So you have now replaced the major components and confirmed that the passive backplane that connected it all together is in one piece.
With a non-dimension cube I have seen the cube shut off the moment you unplugged the monitor. You can hack it to run without a monitor but it requires one hell of a resistor.
It's been a while but if I recall when you press the power button it latched the POWER_ON pin and started up the PSU. Unplugging the monitor removes power from the latch, the POWER_ON pin goes low and the PSU shuts off.
With a dimension cube since the color monitor goes to the dimension board the sound box plugs into the monitor port and does all the original functions of the monochrome megapixel monitor sans the CRT assembly. The assumption is just like in the monitor if you unplug the soundbox the computer shuts off.
You however stated that shorting pin 6 did NOT bring the system up so the problem is neither in the monitor cabling, monochrome monitor or sound box. You also replaced the motherboard which rules out two boards doing the exact same thing. You also replaced the power supply with a supposedly good unit.
Somewhere between the monitor port and the PSU the POWER_ON signal is being lost.
Ok, so I'll start with the bad news. A new backplane doesn't fix the problem. A few interesting observations. My original backplace was copyright 1987 and the new one is 1989. Also there are additional jumpers on the board next to VCC. Gtnicol and Mikeboss, I think you are correct about the cube not drawing enough power. So here is where things get interesting. I resolved myself to not purchase a 3rd power supply. This left me with 2 options. 1) I could try to simulate the load of the monitor, or 2) I could adapt an atx power supply for use with the cube. I chose the latter. I have an image gallery of the cube booting here:
http://www.asterontech.com/Asterontech/next/I have not quite figured out soft booting yet, but I'll post a more detailed explanation of everything when I do. I just got excited and wanted to share/thank everyone for your help so far.
Wow! that is really cool. Nice work. It looks like you could probably swap the guts of that PSU with the NeXT PSU.
That's pretty nifty. FWIW. You can connect a big power resistor to the power supply (between pins 12 and 13) to draw more power if need be. It has to be a BIG resistor though because it needs to draw the equivalent of a monitor!
The advantage to the resistor route would be that it's reversible and you could still use a next monitor with the machine. With my route, the next will no longer be capable of driving a monitor as well. In fact I have not connected the -12V rails at all. I think with this hardware becoming more scarce every day that it will be nice to have ATX as an option going forward. Looking at the inside of the original PSU it became apparent that it was really only a matter of time before it died anyways. I am hoping to find time tonight to work out the soft power situation. At the moment I'm favoring using a cheap uC to sense the ~3v coming from the battery at startup to pull the 'power on' signal low for the ATX PSU. I will post a complete-ish pinout of everything when I'm done. There are a few pins that I will not be able to source without an original working setup. I don't believe they will be necessary for my conversion though.
I used an arduino, and it looks like I solved the soft power issue. I just use an analog pin to check for a voltage level higher or lower than 2v. I have some spare ATMega uC's around so I think I'll just stick with the arduino firmware and create a custom board. It was cool to watch the next startup and shutdown from the keyboard for the first time. More updates to come.
With the work progressing on the emulation route, updated libraries, and inspired hardware mods like this, we can all hope for many years of pleasure from our black hardware and NS/OS.
Nice work barcher174!
OK, as promised I have started a document to list the pinouts and instructions. Consider this a very rough draft. I wanted to get this going before i got lazy.
http://asterontech.com/Asterontech/Next_ATX_Full/
Quote from: "barcher174"OK, as promised I have started a document to list the pinouts and instructions. Consider this a very rough draft. I wanted to get this going before i got lazy.
http://asterontech.com/Asterontech/Next_ATX_Full/
Hello Brian: Amazing work with the ATX board now if you can you only connect it to a pentium board running openstep through the backplain for a dual Cube Computer with original 040 and pentium that would rock!
So, I tested it here with another 152 power supply and the custom cable a flat panel and using an 030 motherboard, yes to confirm it clicks on and off .... I wish I had seen this thread earlier as I honestly did not know the 152 was the cause of the problem as you purchased an 030 board and its power supply and would have sent you a 983 had I known that the cable wouldn't work by default or sold you a NeXT monitor!
Your amazing effort to fix the problem deserves kudos and as
I did not know this was a problem but to make sure I tested everything here as well ... using a the custom cable and 030 motherboard with 152 powersupply yes confirmed not working,
The 983 power supply powers up also you would need this to power up a dimension board .... interestingly the 030 board I'm using is a very early 1988 NeXT gold board and it only puts computer charactors up on the lcd screen not ascii using the 983 supply yet booting with a megapixel monitor it works .
An 040 board works great with a 983 power supply and I'll try a later rev 030 board to see if it work with 983's.
I also tested the custom cable and lcd with sound box tested and work with mono slabs, turbo slabs and adb mono slabs with both adb and non adb components and a Turbo cube motherboard with 2 dimension setup using a 983 and triple 20 inch monitors.
Best regards Rob Blessin :idea:
Thanks Rob. You've been more than helpful. Your cable is well built and I never had reason to suspect it. I was a bit liberal with ordering parts, because I knew I was going to need some spares for some upcoming project ideas. Sometimes you have to crack a few eggs...
As for the current project:
I have successfully moved from the arduino dev board to a bare uC and a few passive components. Everything works as expected from a breadboard. I am now waiting on a few parts so that I can move everything over to a single circuit board. I have settled on gutting one of the original power supplies and will put the ATX PSU inside of it with plenty of room to spare. I'm not a fan of destroying original hardware, but I don't have the resources or the ability to fabricate a replacement housing. If all goes according to plan I should have everything complete this weekend or next. I will update the project log then, and maybe post a youtube video of the whole setup working if there is interest in that sort of thing.
FWIW. I had a bunch of mono->VGA cables built that allow a next box with a soundbox to talk to a VGA/LCD. I had them made... last time I asked, the price would be something like $70 a cable, maybe cheaper if we ordered a large lot.
Quote from: "gtnicol"FWIW. I had a bunch of mono->VGA cables built that allow a next box with a soundbox to talk to a VGA/LCD. I had them made... last time I asked, the price would be something like $70 a cable, maybe cheaper if we ordered a large lot.
If you have any cables remaining I could use one, if not I would be interested in a cable if you were to have more made.
I'll get the cables priced out. Of course, the ideal thing would be to build a replacement for the sound box with video out.
Ok, final update for the project:
http://asterontech.com/Asterontech/Next_ATX_Conversion_REV1/Everything is assembled and working great. I have a molex extension cable on the way so I can power more devices (currently just 1). I now have the backplane powered completely by the ATX connector. You really can't tell anything has been done besides the difference in the molex connectors themselves. I'm overall very pleased with how it turned out. I don't mind if anyone wants to mirror the content. Thanks for the help guys! I have a couple more mods in mind that I think you'll like.
